Suitable methods for landscape evaluation and valorization: the third dimension in landscape metrics

(1)

This article was downloaded by: [Teresa Batista] On: 03 August 2012, At: 09:13

Publisher: Taylor & Francis

Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Acta Botanica Gallica

Publication details, including instructions for authors and subscription information:

http://www.tandfonline.com/loi/tabg20

Suitable methods for landscape evaluation and

valorization: the third dimension in landscape

metrics

Teresa Batista ab , Paula Mendes a , Luisa Carvalho b , Carlos Vila-Viçosa a & Carlos Pinto Gomes a

a

Department of Landscape, Environment and Management/Institute of Environmental and Agricultural Mediterranic Sciences (ICAAM), University of Évora (Portugal), Rua Romão Ramalho, n° 59, P-7000-671, Évora, Portugal

b

Intermunicipal Community of Central Alentejo (CIMAC), Rua 24 de Julho, n° 1, 7000-673, Évora, Portugal

Version of record first published: 02 Aug 2012

To cite this article: Teresa Batista, Paula Mendes, Luisa Carvalho, Carlos Vila-Viçosa & Carlos Pinto Gomes (2012): Suitable methods for landscape evaluation and valorization: the third dimension in landscape metrics, Acta Botanica Gallica, 159:2, 161-168

To link to this article: http://dx.doi.org/10.1080/12538078.2012.696930

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

(2)

Suitable methods for landscape evaluation and valorization: the third dimension in landscape

metrics

Des méthodes appropriées pour l’évaluation et la valorisation du paysage: la troisième

dimension dans les métriques du paysage

Teresa Batistaa,b*, Paula Mendesa, Luisa Carvalhob, Carlos Vila-Viçosaaand Carlos Pinto Gomesa

a

Department of Landscape, Environment and Management/Institute of Environmental and Agricultural Mediterranic Sciences (ICAAM), University of Évora (Portugal), Rua Romão Ramalho, n° 59, P-7000-671 Évora, Portugal;bIntermunicipal Community of Central Alentejo (CIMAC), Rua 24 de Julho, n° 1, 7000-673, Évora, Portugal

Abstract: Landscape metrics have been widely developed over the last two decades. One of the major recent develop-ments in landscape metrics analysis was the integration of the third dimension. Topography has an extremely important role in ecosystems function and structure, even though the common analysis in landscape ecology only considers a plani-metric surface, which leads to some erroneous results particularly in mountain areas. In this study we tested landscape metrics behaviour in 13 sample areas of 10,000 m2_{each in several topographical conditions of Central Alentejo,}

Portu-gal. The signiﬁcance analysis of the results achieved in planimetric and three-dimensional environments is presented. Keywords: Alentejo; landscape metrics; local landscape units; OTALEX II; three dimensions; topography

Resumé: Les métriques du paysage ont été largement développées au cours des deux dernières décennies. Un des récents développements dans le paysage paramètres d’analyse a été l’intégration troisième dimension. La topographie a un rôle extrêmement important sur la fonction des écosystèmes et sur sa structure, même si la commune analyse d ’écolo-gie du paysage ne conçoit que la dimension planimétrique, qui peut conduit à des résultats erronés, en particulier dans les zones de montagne. Dans cette étude nous avons testé paysage métriques, et son comportement en 13 zones de l’échantillon avec des 10.000 m² chacun et dans plusieurs conditions topographiques au Centre de l’Alentejo, au Portu-gal. Il est présenté l’analyse de la signiﬁance des résultats obtenus dans les environnements en 3D et en planimétrie. Mots-clés: 3D; Alentejo; Métriques du paysage; OTALEX II; Topographie; Unités Locales du Paysage

Introduction

Landscape Ecology studies landscape structure, func-tions and changes. Landscape structure is characterized by the composition and configuration of landscape pat-terns. One of the main premises is that landscape structure is connected with landscape functions and processes (von Drop and Opdam 1987; Turner 1989; McIntyre and Wiens 1999). Topography is actually a key factor for many ecological processes, such as ero-sion, flow direction and accumulation, temperature and biodiversity distribution and fire propagation (Swanson et al. 1988; Burnett et al. 1998; Bolstad, Swank and Vose 1998; Davis and Goetz 1990; Blaschke, Tiede and Heurisch 2004). The importance of three-dimen-sional (3D) aspects in landscape ecology was already declared by Carl Troll, at the end of the 1930s, who “hoped that a new science could be developed that could combine the spatial ‘horizontal’ approach of

geographers with the functional ‘vertical’ approach of ecologists” (Farina 1998).

In the past 10 years, 3D issues in landscape ecology have been studied and applied by several researchers using many different approaches (MacNab 1992; Pike 2000; Dorner, Lertzman and Fall 2002; Lefsky et al. 2002; Bowden et al. 2003; Jenness 2004; Sebastiá 2004; McGarigal, Tagil and Cushman 2009). However, most of the landscape ecological studies use the patch–corridor– matrix model to describe the spatial arrangement of land-scape mosaics and patches. This model, used to calculate landscape metrics, is based on area, perimeter and dis-tance planimetric measurements, which can lead to erro-neous results, especially in mountainous areas (Hoechstetter et al. 2008).

Recent landscape ecological studies applied 3D to land-scape metrics (Hobson, 1972; Jenness 2004, 2010; Hoech-stetter, Xuan Thinh and Walz 2006; Hoechstetter et al.

*Corresponding author. Email: tbatista@cimac.pt –168

Société botanique de France

ISSN 1253-8078 print/ISSN 2166-3408 online CopyrightÓ 2012 Société botanique de France http://dx.doi.org/10.1080/12538078.2012.696930 http://www.tandfonline.com

(3)

2008; Sang, Miller and Ode 2008; Hoechstetter 2009; Walz et al. 2010). The study developed by Hoechstetter et al. (2008) provides basic approaches to include relief proper-ties into large-scale landscape analyses, including the calculation of standard landscape metrics on the basis of “true” surface geometries and the application of roughness parameters derived from surface metrology. Others issues when applying 3D, like viewshed to landscape preferences, have been studied by Sang, Miller and Ode (2008).

The aim of the present study was to analyse whether the third dimension is significant in the characterization of a landscape, when based on landscape metrics, as it has already been determined that there are significant dif-ferences between patch, class and landscape metrics cal-culated in planimetric or in 3D environments (Batista, Mendes, Carvalho 2010). It is also tested whether there are significant differences between the sample areas, according to the roughness of the underlying terrain.

This work was carried out by the Environmental Indicators Working Group (EIWG) of OTALEX– Alent-ejo Extremadura Territorial Observatory (www.ideotalex. eu) (in the OTALEX II Project co-ﬁnanced by Opera-tional Programme for Cooperation between cross-border regions of Spain and Portugal – POCTEP), in Central Alentejo (Portugal).

Material and methods Characterization of study area

The study area is located in Central Alentejo, in southern Portugal. It covers about 7400 km2 and has about 175,000 inhabitants, concentrated in small and

medium-sized villages and cities. Altitude varies between 7 and 648 m. We selected 13 sample areas, of 100 km2 each, located along Central Alentejo, representing about 17% of the total area (Figure 1).

Local Landscape Units

To characterize the landscape in the study area, a map was produced using the Local Landscape Units (LLU) based on the methodology proposed by Batista et al. (forthcoming), which integrates Corinne Land Cover level 5 (CLC N5) land cover map at scale 1: 10,000 (Batista, forthcoming), altimetry (Digital Elevation Model (DEM) 25 m), geology and soil units (at scale 1: 25.000). The land cover map applies the hierarchical CLC N5 legend developed by Guiomar et al. (2006, 2009), which has 295 land cover classes. This map was expanded using photo interpretation of digital ortopho-tomaps from 2005 (DGRF, 2005) and field validation at the end of 2008. The Land Cover map was previously generalized to create the LLU map. The relief was ana-lysed using a DEM of 25 25 m pixel. The DEM was reclassified into three altimetry classes according to area roughness (class 1, 0–200 m; class 2, 200–400 m and class 3, 400–655 m). The geology applies the Geo-logic and Miner Portuguese Institute Map, which has 71 geological classes at scale 1: 50,000. To create the LLU map, the geological map was reclassified into four classes, according to their most important geological substrate – Sand, Limestone, Volcanic Rocks (such as granites, granodiorites and tonalities) and Metamorphic rocks (such as schists and greywacke). The soils map was developed by the Portuguese Institute of Hydraulic,

Figure 1. Sample areas localization. Central Alentejo– Portugal.

Figure 1. Localisation de las Zones d’échantillons – Alentejo Central – Portugal.

162 T. Batista et al.

(4)

Rural Engineering and Environment and has 71 com-bined classes at scale 1: 25,000 that were aggregated into 12 main soil classes, following Carvalho Cardoso (1965). From the overlay of these four maps, 103 LLU classes were derived that can be observed in Figure 2. True surface area and perimeter calculation and landscape metrics

Application of 3D to landscape metrics allows us to calculate using true surface area and perimeter measurements (Hoechstetter 2009). Surface area provides a better estimate of the available land area than planimet-ric area, and the ratio of this surface area to planimetplanimet-ric area provides a useful measure of the topographic rough-ness of the landscape (Jenrough-ness 2004).

We used LANDMETRICS-3D, developed by Walz et al.

(2010). LANDMETRICS-3D is an ARCGIS extension that

integrates the available tools for calculating true surface area developed by Jenness (2004, 2010) (http://www.

jennessent.com/arcgis/surface_area.htm, last modiﬁed 8 April 2010) and the fragstats landscape metrics of McGari-gal et al. (2002). This application is based on Jenness (2004), whose technique uses a moving window algorithm and estimates the true surface area for each grid cell using a triangulation method (Figure 3). Each of the triangles is located in three-dimensional space and connects the focal cell with the centre points of adjacent cells. The lengths of the triangle sides and the area of each triangle can be calculated by means of the Pythagorean theorem. The eight resulting triangles are summed to produce the total surface area of the underlying cell (for details see Hoech-stetter et al. 2008; HoechHoech-stetter 2009; Jenness 2010).

LANDMETRICS-3D permits calculation not only of the

true surface and perimeter of the cell, but also of each patch, so as to calculate the landscape metrics. It uses two raster ﬁles of equal resolution (25 25 m) and extent (10,000 10,000 m): the elevation model and the LLU raster. Also the surface distance is calculated.

Figure 2. Sample areas local landscape units.

Figure 2. Zones d’échantillons des unités locales du paysage.

(5)

The following landscape metrics, available in LANDMETRICS-3D, were calculated for the 13 sample

areas, for 2D and 3D: Area (Area), Perimeter (Perim), Fractal Dimension (Fractdim), Perimeter/Area Racio (Ratio), Shape Index (Shape), Total Edge (TEdge), Edge Density (EdgeD), Edge Contrast (EdgeContrast), Largest Patch Index (LPidx), number of patches (PatchNr), Average Roughness (Avg. Roughness), RMS

Roughness (RMS Roughness), Shannon Diversity

Index (ShannonDivInd), Shannon Evenness Index (ShannonEvenInd), Simpson Diversity Index (Simpson-DivInd), Efective Mesh Size (EffectiveMeshSize), Skewness and Kurtosis (for more details of the metrics formulae see Hoechstetter 2009 or MacGarigal and Marks 2004).

Results and discussion

Table 1 presents the results for the landscape metrics as well as the minimum and maximum elevations of each sample area. To test the similarity of the metrics for 2D and 3D, the non-parametric Wilcoxon test was used. The results (Table 2) indicate that all the studied landscape metrics were signiﬁcantly different (p < 0.05) when calculated planimetrically or with the use of the true surface area, perimeter or distance. The only non-signiﬁcant difference was PatchNr, which was not variable across the 2D and 3D and was used as the “blank”. This result support the results achieved in our previous work when we analysed 221,382 records, for the two dimensions (2D and 3D), applied to patch, class and landscape metrics in 11 sample areas and nine landscape metrics (Batista, Mendes and Carvalho 2010).

The range of variation for each landscape metric is also presented, calculated based on the variation in percentage of the 3D landscape metrics in relation to the calculation in the planimetric environment (Figure 4). Both base met-rics Area and Perim show visible variation, more in Area (0.04%) than in Perim (0.01%) (Figure 4). These varia-tions induce the positive variation of EffectiveMeshSize, EdgeContrast, TEdge and diversity metrics ShannonDiv-Ind, ShannonEvenInd and SimpsonDivInd. We observed negative variations in Ratio, EdgeD, LPidx and Shape (very slightly in Shape). This can be explained by the fact that all of these metrics are divided by the Area, which increases in 3D calculations.

FractDim and Shape are the most stable metrics, because they reﬂect the relations between Perim and Area.

The surface metrology metrics Avg. Roughness, RMS Roughness, Skewness and Kurtosis are also presented in the Table 1. Avg. Roughness approximates surface rough-ness by calculating the mean absolute departure of the ele-vation values from the mean plane in m and RMS Roughness is a modiﬁcation of Ra, used as an equivalent

Figure 3. Method to determine true surface area and true surface perimeter of patches. (Figure redrawn according to Jenness 2004 by Hoechstetter et al. 2008).

Figure 3. Méthode pour déterminer l’area réelle de la surface et le périmètre réelle de la surface des taches. (Figure redessiné selon Jenness 2004 par Hoechstetter et al. 2008).

Figure 4. Variation (%) of the three-dimensional landscape metrics in relation to the planimetric calculated metrics. Figure 4. Variation (en%) des 3D métriques en relation avec les 2D métriques.

(6)

T able 1. Landscape metrics results for the 13 sample areas. T ableau 1. Résultats des metriques du Paysage aux 13 zones de l’ echantillon. Metrica A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A 1 1 A12 A13 2D Met_Area_LSC 100500625,00 101252500,00 103016875,00 101757500,00 104295000,00 102000000,00 101757500,00 100747500,00 101505625,00 1 0764 0625,00 1 01505000,00 1 01505000,00 1 03016875,00 3D Met_Area_LSC 101268056,57 104259222,12 107038784,33 104877674,06 106885187,56 105050744,39 105008686,40 102764880,46 101822603,17 1 0778 7065,19 1 02655444,21 1 03186957,43 1 07038784,33 2D Met_Perim_LSC 1403850,00 1675100,00 1670850,00 2342600,00 1974150,00 1886000,00 1497400,00 1 457250,00 1 663450,00 1544600,00 1201650,00 1 7 75550,00 1670850,00 3D Met_Perim_LSC 1409686,55 1690485,71 1674091,22 2364641,10 1991765,14 1902349,60 1514860,92 1 466812,50 1 666046,98 1545814,50 1202662,33 1 7 77130,37 1674091,22 2D Met_ Fract LSC 1,0815 1,0843 1,0659 1,0767 1,0716 1,0825 1,0651 1 ,0717 1 ,0689 1,0647 1,0767 1 ,0775 1,0659 3D Met_ Fract_LSC 1,0815 1,0843 1,0658 1,0763 1,0714 1,0824 1,0653 1 ,0715 1 ,0688 1,0646 1,0765 1 ,0774 1,0658 2D Met_ Ratio_LSC 0,0854 0,0894 0,0952 0,0831 0,0862 0,0716 0,0936 0 ,0836 0 ,0892 0,0953 0,0757 0 ,0781 0,0952 3D Met_ Ratio_LSC 0,0850 0,0886 0,0951 0,0822 0,0850 0,0707 0,0929 0 ,0830 0 ,0890 0,0952 0,0756 0 ,0781 0,0951 2D Met_Shape_LSC 1,6417 1,6492 1,4742 1,5716 1,5398 1,6277 1,4900 1 ,5329 1 ,5055 1,4602 1,6319 1 ,5968 1,4742 3D Met_Shape_LSC 1,6418 1,6482 1,4729 1,5673 1,5358 1,6275 1,4918 1 ,5317 1 ,5053 1,4601 1,6282 1 ,5946 1,4729 2D Met_TEdge_LSC 721975,00 857675,00 855725,00 1 191475,00 1007500,00 963200,00 768875,00 7 48700,00 8 51875,00 793050,00 620975,00 9 07925,00 85 5725,00 3D Met_TEdge_LSC 724942,48 865656,77 857744,22 1202750,44 1016528,48 971638,27 777859,68 7 53679,41 8 53199,04 793667,85 621576,26 9 08853,75 85 7744,22 2D Met_EdgeD_LSC 0,0072 0,0085 0,0083 0,01 17 0,0097 0,0094 0,0076 0 ,0074 0 ,0084 0,0074 0,0061 0 ,0089 0,0083 3D Met_EdgeD_LSC 0,0072 0,0083 0,0080 0,01 15 0,0095 0,0092 0,0074 0 ,0073 0 ,0084 0,0074 0,0061 0 ,0088 0,0080 2D Met_LPidx_LSC 19,6790 15,8132 18,9168 4,1858 5,5731 4,2659 9,5276 1 1,3948 1 1 ,5320 9,5224 10,1325 19,8685 1 8,9168 3D Met_ LPidx LSC 1 9,5919 15,5857 18,2289 4,0770 5,4577 4,1595 9,2350 1 1,2017 1 1 ,5451 9,5192 10,0331 19,5893 1 8,2289 2D Met_EdgeContras t_LSC 9,7629 5,8015 6,3137 4,5860 5,7856 5,5754 5,7626 6 ,7918 7 ,7159 9,1638 14,5266 7,9404 6,3137 3D Met_EdgeContras t_LSC 9,7730 5,8055 6,3559 4,5867 5,7824 5,5819 5,9429 6 ,7894 7 ,7170 9,1641 14,5389 7,9539 6,3559 2D Met_PatchNr_LSC 827,00 1055,00 131 1,00 1769,00 1421,00 1 141,00 1003,00 9 94,00 1 257,00 1231,00 5 49,00 1 1 1 0,00 131 1,00 3D Met_PatchNr_LSC 827,00 1055,00 131 1,00 1769,00 1421,00 1 141,00 1003,00 9 94,00 1 257,00 1231,00 5 49,00 1 1 1 0,00 131 1,00 2D Met_ShannonDivInd _LSC 2,4832 2,7025 2,3672 3,2216 2,9404 3,3762 2,7878 3 ,4333 2 ,7469 2,9046 1,9757 2 ,8106 2,3672 3D Met_ShannonDivInd _LSC 2,4829 2,7231 2,4047 3,2398 2,9628 3,3865 2,8032 3,441 1 2,7466 2,9047 1,9996 2 ,8318 2,4047 2D Met_ShannonEven Ind_LSC 0,6644 0,6944 0,61 15 0,7195 0,6876 0,7772 0,6380 0 ,7688 0 ,6824 0,7123 0,5557 0 ,6661 0,61 15 3D Met_ShannonEven Ind_LSC 0,6643 0,6997 0,6212 0,7236 0,6928 0,7796 0,6415 0 ,7705 0 ,6823 0,7124 0,5624 0,671 1 0,6212 2D Met_SimpsonDivIn d_LSC 0,8838 0,9026 0,8277 0,9337 0,9101 0,9508 0,8983 0 ,9452 0 ,9013 0,9067 0,8051 0 ,8878 0,8277 3D Met_SimpsonDivIn d_LSC 0,8839 0,9049 0,8373 0,9351 0,91 19 0,9518 0,9009 0 ,9462 0 ,9013 0,9067 0,8086 0 ,8906 0,8373 2D Met_EffectiveMeshSize_LSC 6386324,59 4416584,37 7376060,43 795199,73 1204267,66 995949,98 3632484,40 2 905357,08 3 218919,20 2956715,43 42 90210,97 5090375,16 7376060,43 3D Met_EffectiveMeshSize_LSC 6415045,76 4430565,32 7377231,76 81 1302,62 1205333,58 1054004,86 3727508,59 2 908962,74 3 232066,72 2959040,65 4 277989,96 5073026,14 7377231,76 A vg. Roughness 3,9078 2,7566 1,8182 2,5502 2,5434 2,6603 2,9892 2 ,3410 1 ,8805 1,1 140 1,9696 1 ,6474 1,8182 RMS Roughness 4,6198 3,2676 2,1866 3,0322 3,0252 3,1889 3,8107 2 ,7754 2 ,2249 1,3161 2,3476 1 ,9615 2,1866 Skewness 0,0138 0,0048 0,0047 -0,0075 -0,0460 -0,0074 -0,0069 0,0304 0,0598 -0,0222 -0,0472 -0,0151 0,0047 Kurtosis 1,6866 1,6737 1,4725 1,7241 1,6550 1,9564 1,9755 1 ,6576 1 ,6032 1,4632 1,8769 1 ,8159 1,4725

(7)

of the sample standard deviation in statistical analysis. These two metrics provide information about the

rough-ness of the terrain. Figure 5 shows that area 1 (A1 = 3.91 m) is the area that presents more roughness, and area 10

Figure 5. Surface metrology metrics in the 13 sample areas. Figure 5. Surface métrologie métriques aux 13 zones d’échantillon. Table 2. Wilcoxon signed ranks statistics test.

Tableau 2. Wilcoxon signalee statistique.

Landscape Metrics 3D versus 2D Z Asymp. Sig. (two-tailed) 3D Met_Area_LSC - 2D Met_Area_LSC –3.181a 0.001 3D Met_Perim_LSC - 2D Met_Perim_LSC –3.181a 0.001 3D Met_Fract_LSC - 2D Met_Fract_LSC –2.342b 0.019 3D Met_Ratio_LSC - 2D Met_Ratio_LSC –3.181b 0.001 3D Met_Shape_LSC - 2D Met_Shape_LSC –2.412b 0.016 3D Met_TEdge_LSC - 2D Met_TEdge_LSC –3.181a 0.001 3D Met_EdgeD_LSC - 2D Met_EdgeD_LSC –3.181b 0.001 3D Met_LPidx_LSC - 2D Met_LPidx_LSC –3.041b 0.002 3D Met_PatchRD_LSC - 2D Met_PatchRD_LSC –3.181b 0.001 3D Met_PatchNr_LSC - 2D Met_PatchNr_LSC 0.000c _1.000 3D Met_ShannonDivInd_LSC - 2D Met_ShannonDivInd_LSC –2.831a _0.005 3D Met_ShannonEvenInd_LSC - 2D Met_ShannonEvenInd_LSC –2.831a 0.005 3D Met_SimpsonDivInd_LSC - 2D Met_SimpsonDivInd_LSC –3.041a 0.002 3D Met_EffectiveMeshSize_LSC - 2D Met_EffectiveMeshSize_LSC –2.062a 0.039 3D Met_EdgeContrast_LSC - 2D Met_EdgeContrast_LSC –2.552a 0.011

a. Based on negative ranks. b. Based on positive ranks.

c. The sum of negative ranks equals the sum of positive ranks. 166 T. Batista et al.

(8)

(A10 = 1.11 m) shows less roughness. All the remaining values varied between 1.5 and 3 m. Hence none of the present sample areas had high roughness, but they still presented signiﬁcant differences between the metrics in 2D and 3D.

Skewness varied from – 0.05 (A11) to 0.06 (A9), but was very close to zero, which indicates a symmetrical shape for the surface height distribution on average. Skew-ness may be negative if the distribution has a longer tail at the lower side of the mean plane, or positive if the distri-bution has a longer tail at the upper side of the mean plane. As a consequence, it can give some indication of the exis-tence of“spiky” features (Hoechstetter 2009).

Kurtosis varies between 1.46 (A3 and A13) and 1.98 (A9), which indicates a well spread distribution of the surface height, in every sample area.

Conclusion

This study supports the thesis that use of the third dimension can be important in landscape metrics calcula-tions. The results achieved here reveal that all landscape metrics have signiﬁcant differences when calculated in 3D instead of planimetrically. Only two of the studied metrics, FractDim and Shape, show almost no difference between 2D and 3D, but the Shape metric presents a very slight negative difference. These results agree par-tially with the results achieved by Hoechstetter (2009), but our sample areas have low roughness, symmetrical shape for the surface height distribution and well spread distribution of the surface height, which indicates that, even in areas with less relief, the use of the real surface metrics can make a difference in landscape analysis. LANDMETRICS-3D was shown to be a very useful tool for

landscape analysis in 3D and should its development should continue.

References

Batista, T. forthcoming. Carta de ocupação/uso do solo do Alentejo Central e Concelho de Sousel – Legenda Corine Land Cover Nível 5. Évora: CIMAC Edts.

Batista, T., P. Mendes, and L. Carvalho. 2010. The third dimen-sion in landscape metrics analysis applied to central Alente-jo – Portugal. Proceedings of the IUFRO Landscape Ecology Working Group International Conference, Sept. 21–27 2010, IUFRO. Bragança: Portugal, 462–7.

Blaschke, T., D. Tiede, and M. Heurich. 2004. 3D-landscape metrics to modelling forest structure and diversity based on laser-scanning data. In Proceedings of the ISPRS working group VIII/2, “Laser-Scanners for Forest and Landscape Assessment” ed. M. Thies, B. Koch, H. Spiecker and H. Weinacker. International Society of Photogrammetry and Remote Sensing 129–132.

Bolstad, P.V., W. Swank, and J. Vose. 1998. Predicting South-ern Appalachian overstory vegetation with digital terrain data. Landscape Ecology 13: 271–283.

Bowden, D.C., G.C. White, A.B. Franklin, and J.L. Ganey. 2003. Estimating population size with correlated sampling unit estimates. Journal of Wildlife Management 67: 1–10.

Burnett, M.R., P.V. August, J.H. Brown, Jr., and K.T. Killing-beck. 1998. The inﬂuence of geomorphological heterogene-ity on biodiversity: I. A patch scale perspective. Conservation Biology 12: 363–370.

Carvalho Cardoso, J. 1965. Os solos de Portugal. Sua classi ﬁ-cação, caracterização e génese, I. A sul do rio Tejo. Direc-ção-Geral dos Serviços Agrícolas, Lisboa, 331 pp.

Davis, F.W., and S. Goetz. 1990. Modeling vegetation pattern using digital terrain data. Landscape Ecology 4: 69–80. Dorner, B., K. Lertzman, and J. Fall. 2002. Landscape pattern

in topographically complex landscapes: issues and tech-niques for analysis. Landscape Ecology 17: 729–743. DGRF. 2005. V Inventário Florestal Nacional: Manual de

Instruções para a realização do Trabalho de Campo. Direc-ção Geral dos Recursos Florestais: Lisboa.

Farina, A. 1998. Principles and methods in landscape ecology. London: Chapman & Hall. 235 pp.

Guiomar, N., T. Batista, J.P. Fernandes and C.S. Cruz. 2009. Corine Land Cover Nível 5 – Contributo para a carta de Uso do Solo em Portugal Continental. Évora (Portugal): AMDE Edts. 226 pp.

Guiomar, N., J.P. Fernandes, C.S. Cruz, T. Batista, and J. Mateus. 2006. Sistemas de classiﬁcação e caracterização do uso e ocupação do solo para zonamento microescalar. Pressupostos para a adaptação da Legenda CORINE Land Cover (Nível 5) à escala 1:10000 e análise comparativa de sistemas de classiﬁcação de uso e ocupação do solo. Pro-ceedings ESIG2006 15–17. Taguspark, Oeiras, Portugal. Hobson, R.D. 1972. Chapter 8 Surface roughness in topography:

quantitative approach. In Spatial analysis in geomorphology, ed. R.J. Chorley, 221–45. New York: Harper & Row. Hoechstetter, S. 2009. Enhanced methods for analysing

landscape structure. Landscape metrics for characterising three-dimensional patterns and ecological gradients. Band 6 der Reihe, Fernerkundung und angewandte Geoinforma-tik. Berlin: Rhombos-Verlag.

Hoechstetter, S., U. Walz, L.H. Dang, and N.X. Thinh. 2008. Effects of topography and surface roughness in analyses of landscape structure – a proposal to modify the existing set of landscape metrics. Landscape Online 3: 1–14.

Hoechstetter, S., N. Xuan Thinh, and U. Walz. 2006. 3D-Indices for the Analysis of Spatial Patterns of Landscape Structure. Proceedings, InterCarto– InterGIS 12. Berlin 108–18. Jenness, J.S. 2004. Calculating landscape surface area from

digital elevation models. Wildlife Society Bulletin 32, no. 3: 829–839.

Jenness, J. 2010. http://www.jennessent.com/arcgis/surface_area. htm/ (Surface Area and Ratio for ArcGIS (surface_area.exe) v. 1.0.137. Jenness Enterprises). Accessed in March 2010. Lefsky, M.A., W.B. Cohen, G.P. Parker, and D.J. Harding.

2002. Lidar remote sensing for ecosystem studies. BioSci-ence 52: 19–30.

McGarigal, K., S.A. Cushman, M.C. Neel, and E. Ene. 2002. FRAGSTATS: Spatial Pattern Analysis Program for Cate-gorical Maps– Manual for the computer software program produced by the authors at the University of Massachusetts, Amherst.

MacGarigal, K., and B.J. Marks. 2004. FRAGSTATS: Spatial Pattern Analysis Program for quantifying landscape struc-ture V. 2.0. Forest Science Departament, Oregon State Uni-versity, Corvallis (OR).

McGarigal, K., S. Tagil, and S.A. Cushman. 2009. Surface met-rics: an alternative to patch metrics for the quantiﬁcation of landscape structure. Landscape Ecology 24: 433–450. McIntyre, N.E., and J.A. Wiens. 1999. Interactions between

landscape structure and animal behavior: the roles of heter-ogeneously distributed resources and food deprivation on movement patterns. Landscape Ecology 14: 437–447.

(9)

McNab, W.H. 1992. A topographic index to quantify the effect of mesoscale landform on site productivity. Canadian Jour-nal of Forest Research 23: 1100–1107.

Pike, R.J. 2000. Geomorphometry - diversity in quantitative surface analysis. Progress in Physical Geography 24: 1–20.

Sang, N., D. Miller, and Ð. Ode. 2008. Landscape metrics and visual topology in the analysis of landscape preference. Environment and Planning B: Planning and Design 35: 504–520.

Sebastiá, M.-T. 2004. Role of topography and soils in grassland structuring at the landscape and community scales. Basic and Applied Ecology 5: 331–346.

Swanson, F.J., T.K. Kratz, N. Caine, and R.G. Woodmansee. 1988. Landform effects on ecosystem patterns and pro-cesses. BioScience 38: 92–98.

Turner, M.G. 1989. Landscape ecology: the effect of pattern on process. Annual Review of Ecology and Systematics 20: 171–197.

van Drop, D., and P.F.M. Opdam. 1987. Effects of patch size, isolation and regional abundance on forest bird communi-ties. Landscape Ecology 1: 59–73.

Walz, U., S. Hoechstetter, D. Vock, and L.H. Dang. 2010. LandMetrics-3D. Landscape metrics for raster-based struc-tureanalysis in three dimensions. Leibniz Institute of Eco-logical and Regional Development (IOER). Weberplatz 1, D-01217 Dresden, Germany. 10 pp.